Nanbeige4.1-3B
- Input
- —
- Output
- —
- Speed
- ~110tok/sest.
- Context
- 256Ktokens
Long-tail import from Artificial Analysis (2026-09-24); AA variant: Nanbeige4.1-3B. Price = AA median host price.
Rankings
Self-hostingEstimated
How we estimate →What it would cost to run these open weights yourself on rented GPUs. No API sells this model, so this is its price on the chart.
- Hardware
- 1× B200 180GB
- BF16 weights
- Throughput
- ~15,920 tok/s
- many requests batched
- Price
- $0.047–0.42 /M tok
- busy → light use
- On your own machine
- RTX 5090 32GB (int4, 300+ tok/s single-stream)
- single consumer GPU or Mac
Assumptions (5)
- BF16 weights (3.93B params) + 25% KV-cache headroom ≈ 10 GB
- 1× B200 180GB at $5.98–$14.24/GPU-hour on-demand (2026-09-24)
- ~15,920 output tok/s aggregate at batch 256 (bandwidth-bound)
- Blended 3:1 input:output; prefill ~135,258 tok/s
- Low = 75% utilization at the low GPU price; high = 20% at the high price
Benchmark scores
~ italic, dashed = no published score yet, estimated from related benchmarks.
Writing & chat
IFBenchVerified
#98 of 13935.4%
- #1MiniMax-M382.9%
- #2Qwen3.8 2.4T A95B82.8%
- #97Mistral Large 336.2%
- #98Nanbeige4.1-3B35.4%
- #99Jamba 1.7 Large35.2%
artificialanalysis.ai · 2026-09-24
MMMLUEstimated
~last of 5 (estimated)below measured range (<81.3%)
No published score — estimated from related benchmarks
LMArena Text, Non-English (Elo)Estimated
~#46 of 57 (estimated)~1344 Elo
- #1Gemini 4 ArgonFew votes1512
- #2Claude Opus 5.51497
- #45Trinity Large Thinking1350
- ~#46Nanbeige4.1-3BEstimated1344
- #46gpt-oss-120b1338
No published score — estimated from related benchmarks
LMArena Text (Elo)Estimated
~#46 of 57 (estimated)~1365 Elo
- #1Gemini 4 Argon1525
- #2Claude Opus 5.51507
- #45Trinity Large Thinking1369
- ~#46Nanbeige4.1-3BEstimated1365
- #46INTELLECT-31356
No published score — estimated from related benchmarks
EQ-Bench Creative Writing v3 (Elo)Estimated
~#22 of 26 (estimated)~1437 Elo
- #1GPT-6 Astra2173
- #2Claude Fable 5.12162
- #21Gemini 3.1 Pro1491
- ~#22Nanbeige4.1-3BEstimated1437
- #22Mistral Large 31412
No published score — estimated from related benchmarks
Hard reasoning
GPQA DiamondVerified
#61 of 19284.9%
- #1GPT-6 Astra96.1%
- #2Gemini 3.8 Flash95.3%
- #61MiMo-V2.584.9%
- #61Nanbeige4.1-3B84.9%
- #63Qwen3.6 35B A3B84.1%
artificialanalysis.ai · 2026-09-24
Humanity's Last ExamVerified
#109 of 19810.9%
- #1Claude Fable 5.274.3%
- #2Claude Opus 5.561.4%
- #109Hermes 4 - Llama-3.1 405B10.9%
- #109Nanbeige4.1-3B10.9%
- #111NVIDIA Nemotron-Labs Dory1.0 4B10.9%
artificialanalysis.ai · 2026-09-24
AIME (latest)Estimated
~#22 of 34 (estimated)~79.8%
- #1GLM-5.3 NVFP499.5%
- #2Beam 1.0 501B A23B97.7%
- #21Magistral Small 1.280.3%
- ~#22Nanbeige4.1-3BEstimated79.8%
- #22Phi-4-reasoning-plus78.0%
No published score — estimated from related benchmarks
AA Intelligence IndexEstimated
~#70 of 103 (estimated)~16.1
- #1Gemini 4 Carbon59.4
- #2Claude Opus 5.557.6
- #69Ring-2.6-1T16.6
- ~#70Nanbeige4.1-3BEstimated16.1
- #70K2 Horizon 3.7B15.6
No published score — estimated from related benchmarks
Agents
τ²-bench (Telecom)Verified
#78 of 11921.6%
- #1Step 3.7 Flash98.5%
- #2DeepSeek-V4-Pro-0813-nvfp4-DSpark98.3%
- #78Hermes 4 - Llama-3.1 70B21.6%
- #78Nanbeige4.1-3B21.6%
- #78Qwen3 Next 80B A3B Instruct21.6%
artificialanalysis.ai · 2026-09-24
OSWorld-VerifiedEstimated
~#12 of 15 (estimated)~56.4%
- #1Kimi K384.8%
- #2Qwen3.8 27B84.3%
- #11Muse Glimmer65.9%
- ~#12Nanbeige4.1-3BEstimated56.4%
- #12Nemotron 3.5 Super VL NVFP452.3%
No published score — estimated from related benchmarks
OSWorld 2.0Estimated
~#3 of 5 (estimated)~53.8%
No published score — estimated from related benchmarks
BrowseCompEstimated
~#21 of 22 (estimated)~40.1%
- #1Atria Dawn Preview92.5%
- #2GPT-6 Astra91.5%
- #20Nemotron 3 Ultra44.4%
- ~#21Nanbeige4.1-3BEstimated40.1%
- #21Mistral Small 421.3%
No published score — estimated from related benchmarks
GDPval-AA (Elo)Estimated
~#38 of 92 (estimated)~1074 Elo
- #1Claude Opus 5.51866
- #2Muse Wonder Full1857
- #37Inkling1076
- ~#38Nanbeige4.1-3BEstimated1074
- #38Grok Build 0.1 06161052
No published score — estimated from related benchmarks
AutomationBench-AAEstimated
~#47 of 85 (estimated)~10.7%
- #1Claude Fable 5.276.4%
- #2Muse Wonder Full71.3%
- #46K2 Horizon 3.7B14.7%
- ~#47Nanbeige4.1-3BEstimated10.7%
- #47Qwen3.5 397B A17B6.9%
No published score — estimated from related benchmarks
τ-Bench Banking (AA)Estimated
~#57 of 91 (estimated)~12.6%
- #1Muse Spark 1.350.5%
- #2GLM-5.350.3%
- #56gpt-oss-120b12.8%
- ~#57Nanbeige4.1-3BEstimated12.6%
- #57Gemma 4 26B A4B12.0%
No published score — estimated from related benchmarks